18 On the Five Universals in General
On the Five Universals in General culty is removed when we note that the argument just presented is conclusive only if the relevant common terms are suppositing uniformly in all the propositions in which they appear. Careful consideration shows that they are not.
Nevertheless, one could use the term 'less general' in a different way. He could argue that one term is less general than another if the second is predicated of the first (along with others) when the first is Suppositing in some way or other. Thus, it might be that the more general term cannot be predicated ofits inferior when the inferior is suppositing in a different way, so that it would not be predicated of that inferior when it supposits in all ways. In this new sense one could hold that the same thing is both more and less general than some other thing, but in the revised usage 'less general' and 'more general' cease to function as opposites. They are simply different notions.
18: On the Five Universals in General Now that we have shown what a universal is we ought to determine how many species of universals there are. There are five species, and this can be shown in the following way. Every universal is predicated of many either in quid or not in quid. If it is predicated in quid, then it can be used as an answer to the question
'What is it?" There are two possible cases here. In the first case the many things of which the universal is predicated are all alike so that they agree essentially, although it can happen that one of them is composed of several such similar things: here we have a lowest level species. In the second case not all the things of which the universal is predicated agree in the aforesaid way, but among them there are things which are dissimilar both as wholes and in their parts. Animal is an example. Animal is predicated of both man and donkey, but the similarity in substance between two men is greater than that obtaining between a man and a donkey. The same is true in the case of color. This term is predicable of both whiteness and blackness, but neither this blackness nor oneofits parts agrees as much with this whiteness or one of its parts as one bit of whiteness agrees with another. For this reason the intention predicable of whiteness and blackness is not a lowest level species but a genus. But whiteness is a lowest level species with respect to all whitenesses. Admittedly, it sometimes happens that one whiteness agrees more with a second whiteness than with a third. Thus, equally intense whitenesses seem to agree more than two whitenesses of differOn the Five Universals in General ent intensities. Nevertheless, given two such whitenesses, one always agrees with some part of the other as much as any two whitenesses agree with each other. For this reason, whiteness is a lowest level species and not a genus with respect to whitenesses.
Nevertheless, it should be noted that both 'genus' and 'species' have two senses, one broad and the other narrow. In the narrow sense we call something a genus if it can be used to answer the question 'What is —?' where the relevant object is referred to by a pronoun. Thus, one could answer the question
'What is this?' (where 'this' refers to Socrates) by saying that he is an animal, that he is a man, and so on with other genera. The same point holds of species.
But in the broad sense something is said to be a genus or species which can be used to answer the question 'What is—?' where the question incorporates some connotative form which is not absolute. Thus, colored seems to answer the question 'What is the white?" However, a what-question whose referential device is a demonstrative pronoun can never be answered by colored; for no matter what we take to be the referent of'this' in 'What is this?' one could never correctly answer that question with colored. If in asking the question one is referring to the subject of whiteness, then it is clear that the question is not correctly answered in that way. If the referent is whiteness, it is likewise clear that one cannot respond by saying "Colored;" for whiteness is not itself colored. If the referent is an aggregate, then again the response "Colored" is inadequate; for as I shall show later, no aggregate is colored. Finally, if the referent be the term
'white', the response is incorrect; for no term is colored. Thus, since one can correctly respond to the question, 'What is the white?' by saying 'Colored", colored can be called a genus in the broad sense of the term. Nevertheless, since colored cannot be used as an answer to a what-question whose referential device is the demonstrative pronoun, it is not a genus in the narrow sense of 'genus'.
The same sort of account holds in the case of species. This distinction is necessary since without it we cannot handle many of the texts of Aristotle and other writers. Using this distinction we can explicate the various rules that hold apropos of genus and species. Many of these must be understood to hold of things that are specics and genera in the first sense. Such rules do not apply in the case of what we call genera and species in the broad sense, and this will become clearer as we go on.
But if a universal is not predicated in quid, it can express one part ofa thing and not another while expressing nothing extrinsic to the thing. In this case we have a difference. If rational, for example, is the difference of man, it expresses a part of man, such as the form, but not, for example, the matter. On the other hand, a universal can express something which is not a part of the thing. In that case it is predicable either necessarily or contingently. If contingently we call it an accident, if necessarily we call it a property.